Multi-Agent Reinforcement Learning on LBF 8x8-3p-2f
95.9Success Rate (SR)EMA
Evaluation Results
| Method | Links | |
|---|---|---|
| EMAalpha=0.22026.07 | 95.9 | |
| Dynamic LLMgranularity=per-episode2026.07 | 95 | |
| Freeze-Schedule2026.07 | 94.4 | |
| Baseline QMIXshaping=no2026.07 | 0.1 |
Loading the SOTA2 catalog…